Breast Cancer
Conditions
Keywords
Mammography
Brief summary
The purpose of this pilot is to provide credible performance estimate information in order to conduct subsequent reader studies.
Detailed description
This pilot reader study was to determine the following: 1. the radiologist's performance metrics for the two reading modalities, i.e., FFDM read on the Aspire Bellus workstation and FFDM read in conjunction with DBT read on the Aspire Bellus workstation,; 2. the magnitude and direction of differences between performance metrics for the two modalities; and 3. variance components and correlations that would influence samples sizes and case mix for the subsequent pivotal reader study comparing performance metrics between the two modalities.
Interventions
DEVICEFFDM Plus DBT
FujiFilm Aspire Cristalle System
DEVICEFFDM
FujiFilm Aspire Cristalle System
Sponsors
Fujifilm Medical Systems USA, Inc.
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
DIAGNOSTIC
Masking
NONE
Eligibility
Sex/Gender
FEMALE
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Female subjects participating in FMSU004A protocol with known clinical status
Exclusion criteria
* Subjects with unknown clinical status not participating in FMSU004A protocol.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Compare Per Subject Area Under Curve (AUC): FFDM Only vs DBT Plus FFDM | 1 month | Breast AUC performance metrics to determine if FFDM plus DBT improved cancer detection rate, requiring correct lesion localization. Statistician to estimate AUCs for each reader in each review condition (FFDM read in conjunction with FFDM plus DBT read) based on their Probability of Malignancy (POM) scores. POM scores will require correct lesion localization, such that in a case with cancer if the reader recorded one or more findings in the case but none of them are determined by the truther to match the location(s) of any proven malignancies, a POM score of 0 will be assigned to the case. Statistician to provide graphical representations of each reader's ROC curve for each review condition. For each reader the difference between the AUC for the FFDM read in conjunction with the FFDM plus DBT will be presented. Statistician to perform MRMC comparison of AUC's between FFDM read in conjunction with FFDM plus DBT using the MRMC method of Dorfman, Berbaum & Metz (1992). |
Countries
United States
Participant flow
Recruitment details
The 100 breast screening and diagnostic cases selected for this study, were obtained from the FMSU2013-004A acquisition study. The statistician randomly selected cases to provide a distribution of density, finding types, etc.
Pre-assignment details
Each radiologist read all 100 cases as FFDM only and FFDM+DBT(100 cases multiplied by 2 modalities = 200 reads) with a 4 week washout period between the 2 sessions.
Participants by arm
| Arm | Count |
|---|---|
| All 100 Cases/Study Participants Each case contributed 2 sets of images: FFDM only as well as DBT plus FFDM. Therefore there were 200 reads acquired from the 100 cases/participants. | 100 |
| Total | 100 |
Baseline characteristics
| Characteristic | All 100 Cases/Study Participants |
|---|---|
| Age, Customized Participant Age | 58 years STANDARD_DEVIATION 10.3 |
| Race/Ethnicity, Customized Asian | 3 Participants |
| Race/Ethnicity, Customized Black or African American | 8 Participants |
| Race/Ethnicity, Customized Other/Unknown | 2 Participants |
| Race/Ethnicity, Customized White | 87 Participants |
| Region of Enrollment United States | 100 Participants |
| Sex: Female, Male Female | 100 Participants |
| Sex: Female, Male Male | 0 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 0 / 100 | 0 / 100 |
| other Total, other adverse events | 0 / 100 | 0 / 100 |
| serious Total, serious adverse events | 0 / 100 | 0 / 100 |
Outcome results
Primary
Compare Per Subject Area Under Curve (AUC): FFDM Only vs DBT Plus FFDM
Breast AUC performance metrics to determine if FFDM plus DBT improved cancer detection rate, requiring correct lesion localization. Statistician to estimate AUCs for each reader in each review condition (FFDM read in conjunction with FFDM plus DBT read) based on their Probability of Malignancy (POM) scores. POM scores will require correct lesion localization, such that in a case with cancer if the reader recorded one or more findings in the case but none of them are determined by the truther to match the location(s) of any proven malignancies, a POM score of 0 will be assigned to the case. Statistician to provide graphical representations of each reader's ROC curve for each review condition. For each reader the difference between the AUC for the FFDM read in conjunction with the FFDM plus DBT will be presented. Statistician to perform MRMC comparison of AUC's between FFDM read in conjunction with FFDM plus DBT using the MRMC method of Dorfman, Berbaum & Metz (1992).
Time frame: 1 month
Population: The study employed a fully factorial, counterbalanced crossover design in which all readers reviewed images from all cases in two visits separated by a memory washout period of approximately 4 weeks. Each reader read half FFDM cases and half FFDM + DBT cases during each of 2 visits.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| FFDM Plus DBT | Compare Per Subject Area Under Curve (AUC): FFDM Only vs DBT Plus FFDM | 100 Probability | Standard Error 0.812 |
| FFDM | Compare Per Subject Area Under Curve (AUC): FFDM Only vs DBT Plus FFDM | 100 Probability | Standard Error 0.78 |